Análise e otimização de linhas de transmissão com elevada capacidade de transmissão de energia usando o método adjunto

Abstract

Three-phase overhead transmission lines (TL's) are fundamental elements of the electrical power system. Because of the demand for the transmission of increasingly large energy blocks, the need to build TL's with high transmission capacity is imminent. Transmission undertakings have faced environmental and economic difficulties in the construction of new energy corridors. Alternatively, there is the development of methodologies for obtaining TL's with high transmission capacity from existing energy corridors. Therefore, energy concessionaires and research centers have been dedicated to the development of these methodologies in the last decades. In this work, a methodology is developed to obtain new conceptions of conductor bundles obtained from minimal changes in the geometries of existing energy corridors and with low electromagnetic imbalances involving the combination of phase compaction, bundle expansion, and high surge impedance loading (HSIL) concepts. For this purpose, an analytical and numerical model of the TL understudy is developed for the determination of the surface and at ground-level electric fields, and the magnetic field involved, and to calcultate the surge impedance loading (SIL) of the LT. Deterministic optimization is performed using the multiple cuts ellipsoid method and the gradient descent algorithm. These methods use the sensitivity of the positions cables concerning the increase in the transmission capacity obtained through the adjoint method. The adjoint sensitivity provides the sensitivity analysis (derived from the objective function concerning the parameter of interest) of the problem, solving only one more linear system of equations quickly and efficiently and regardless of the number of variables considered. Due to the complexity of the analyzed system, this sensitivity can become prohibitive if it is performed using traditional methods, such as the finite difference method. The results show that the optimization based on the adjoint method rise the transmission capacity 20 % using the new bundle geometries suggested by the proposed methodology.